Combined Lagrange Multiplier Test for ARCH in Vector Autoregressive Models

In this paper we propose a combined Lagrange multiplier (LM) test for autoregressive conditional heteroskedastic (ARCH) errors in vector autoregressive (VAR) models by following a suggestion in Dufour et al. (2010) of replacing an exact Monte Carlo (MC) test by a bootstrap MC test when the model includes lags. The test circumvents the problem of high dimensionality in multivariate tests for ARCH in VAR models. It is computationally simple since it only requires computing univariate statistics. The bootstrap MC test is shown to be asymptotically exact. Monte Carlo simulations show that the test has good finite-sample properties. It is found that the test is robust against a non-normal error distribution, while other multivariate LM tests for ARCH suffer from size distortion. We present two financial applications of multivariate LM tests for ARCH to credit default swap (CDS) prices and Euribor interest rates. The results indicate that the errors are skewed and heavy-tailed, and that there are significant ARCH effects.